In most of the optical interferometry techniques, one often get interferometric fringe patterns which include noises, and these noises will cause analysis difficulty. The method of median filter is used to sort the data in the region, and use median value to replace the center value of the region. Therefore, the median filter technique can find central tendency of the region, and less affected by the presence of a minority of aberrant value. So the median filter technique is used frequently in image processing of many optical fringe patterns
The median filter technique is effective in removing noise with high frequency, but tends to smooth out details such as summits if the window size is large. So there are many improved methods developed corresponding to different properties of signal. Even though each of the improved methods has different considerations, such as to consider different integer weights, or different window size etc., their purposes are all the same, that is to make results more accurately. On the basis of these viewpoints, this thesis will use concept of least¡Vsquares fitting on 2-D interferometric fringe patterns. By finding the trend of the region, and avoid unnecessary signal consideration, make the results more accurately.
In order to study the effect of the proposed technique, two sets of fringe patterns obtained from the phase-shifting speckle interferometry will be used as examples. They showed that the proposed technique presents a better result than traditional technique.
| Identifer | oai:union.ndltd.org:NSYSU/oai:NSYSU:etd-0908104-182339 |
| Date | 08 September 2004 |
| Creators | Tsai, Wen |
| Contributors | none, none, Chien, Chi-Hui, none |
| Publisher | NSYSU |
| Source Sets | NSYSU Electronic Thesis and Dissertation Archive |
| Language | Cholon |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | http://etd.lib.nsysu.edu.tw/ETD-db/ETD-search/view_etd?URN=etd-0908104-182339 |
| Rights | unrestricted, Copyright information available at source archive |
Page generated in 0.0023 seconds